Self-exciting point process modeling of conversation event sequences

TL;DR

This paper investigates the properties of Hawkes self-exciting processes in modeling conversation sequences, demonstrating their ability to produce bursty event patterns and applying the model to real-world office conversation data.

Contribution

It provides an analysis of Hawkes process properties, fits the model to conversation data, and discusses limitations in controlling burstiness and interevent time correlations.

Findings

Hawkes processes generate bursty interevent times across parameters.

Model fitting reveals individual differences in self-excitation and decay.

Limitations exist in independently modulating burstiness and interevent correlations.

Abstract

Self-exciting processes of Hawkes type have been used to model various phenomena including earthquakes, neural activities, and views of online videos. Studies of temporal networks have revealed that sequences of social interevent times for individuals are highly bursty. We examine some basic properties of event sequences generated by the Hawkes self-exciting process to show that it generates bursty interevent times for a wide parameter range. Then, we fit the model to the data of conversation sequences recorded in company offices in Japan. In this way, we can estimate relative magnitudes of the self excitement, its temporal decay, and the base event rate independent of the self excitation. These variables highly depend on individuals. We also point out that the Hawkes model has an important limitation that the correlation in the interevent times and the burstiness cannot be independently modulated.